The stress release model and results from modelling features of some seismic regions in China

The linked stress release model for spatio-temporal seismicity: formulations, procedures and applications

“…For details of the data, geology and tectonics we direct the reader to Vere‐Jones & Deng (1988), Zheng & Vere‐Jones (1991, 1994) and Zhuang & Ma (1998). Briefly, the data are considered complete (Gu 1983a,b), where M is the instrumental magnitude estimated from historical evidence of the intensity and extent of shaking.…”

“…Geophysical considerations (Eguchi 1983; Li & Liu 1986) argue for there being two or four distinct regions. Three classifications of events into regions were considered: (A) that of Zheng & Vere‐Jones (1994), (B) that of Zhuang & Ma (1998), which are based on slightly different geophysical interpretations, and (C) via a clustering algorithm using a centroid criterion. The latter is included, even though using the data in this way adds several degrees of freedom to the model, in order to see if there is any measurable effect.…”

“…The stress release model has been applied to historical data from the Kamakura region of Japan (Vere‐Jones 1978), north China (Vere‐Jones & Deng 1988; Zheng & Vere‐Jones 1991, 1994; Zhuang & Ma 1998), southwest China and Taiwan (Zhuang & Ma 1998), central Japan and Persia (Zheng & Vere‐Jones 1994) and southwest Japan (Imoto 2001). The repeated investigations of north China were driven by incremental improvements in the quality of data, and by subtle changes to the identification of regions, which we shall examine later.…”

“…The conclusion was reached that, provided the data are reasonably complete, the stress release model is superior statistically to the Poisson and Poisson with trend models. Zhuang & Ma (1998) also simulated the model forward to forecast a quiescent period of 30–50 yr for the easternmost part of north China.…”

The degree of predictability of earthquakes in several regions of China: statistical analysis of historical data

“…Here, the data from three strong earthquake-prone regions, including North China, Southwest China, and Taiwan, as illustrated in Fig. 1, are chosen in terms of both the completeness and extensiveness of historical records (Gu, 1983;Zheng and Vere-Jones, 1991;Zhuang and Ma, 1998). We should notice that in all the data sets studied below, aftershocks were deliberately removed from earthquake catalogues using the simple window system of Gardner and Knopoff (1974) since the model framework presented above is only suitable to the analysis of mainshocks.…”